"No matter how many communes anybody invents," said Margaret Mead, "the family always creeps back." Family is one of the most important institutions in society and the basic unit of microeconomic analysis. Economists have long been fascinated by decision making behaviour of families. Changes in marriage, fertility rates, divorce rates and social norms in both developed countries and developing countries have given economists an opportunity to understand household decision making behaviour. Traditionally, households have been viewed as a 'black box' having a single utility function. Household behaviour, according to the 'common preference' models, was rationalized as attempting to maximize this single utility function, subject to a budget constraint. Recent studies have challenged this "unitary" approach of household decision-making and have attempted to incorporate divergent or conflicting preferences of individual family members into household decision-making. Empirical studies have shown that household members, who may have different preferences, bargain over the allocation of household resources. The weight given to each persons input depends on his or her income and utility outside the marriage.
Economic models of consumer demand and labour supply have focused on a single
agent and his or her actions. This individualistic theory conflicts with
the reality that people tend to live and work in families, which have multiple
decision makers. Economists have dealt with the multiplicity of decision
makers within a family by proposing two models of household behaviour. The
first approach treats the family as though it were a single entity. I will
discuss the 'common preference' approach to household decision making and
the two models - Samuelson (1956) and Becker (1974, 1981) - that provide
its theoretical underpinnings in Section I. Changes in laws and social norms
have allowed economist to empirically test the 'common preference' and income
pooling hypothesis. I will discuss some such studies and critique the model
and its income pooling assumption in Section II. I will then outline the
alternative bargaining model of household decision-making in Section III
and discuss the two different types of bargaining - divorce threat bargaining
and non-cooperative bargaining. I will discuss some studies that support
the bargaining model of household decision-making in Section IV.
Common Preference
Traditionally, much of the work in economics has treated the household as
a single economic actor. The models - Samuelson (1956) and Becker (1974,
1981) - assume that households have a common preference and, therefore,
a single utility function. Households' single utility function includes
the consumption and leisure time of every family member (Lundberg and Pollak,
1996). Household decision-making, in the 'common preference' models, is
rationalized as the outcome of maximizing a single utility function, subject
to a budget constraint. Since all the income is pooled, only total household
income affects the budget constraint and thus the family demand. Which member
of the family receives the income is irrelevant to intrahousehold allocation.
The assumption of single household utility function does not, however, mean
that all individuals in a household have same or even similar preferences.
In a two-member household consisting of a husband and wife, for example,
each individual has an individual utility function that depends on his or
her private consumption of goods and services. The 'common preference' models
assume that both the husband and wife, who may have different individual
utility functions, agree to maximize a consensus social welfare function
of their individual utilities (Lundberg and Pollak, 1996).
The Samuelson (1956) 'common preference' model, however, did not explain
how the family achieves a consensus for allocating the pooled income to
maximize the joint utility function. Becker (1974, 1981) hypothesized that
the family consists of a group of purely selfish "kids" and one
altruistic parent (Lundberg and Pollak, 1996). The altruistic parents' utility
function reflects a concern for the well-being of other family members.
Becker argued that the presence of an altruistic parent who makes positive
transfers to each member of the family induces the selfish kids to ach in
an apparently unselfish way (Lundberg and Pollak, 1996). The altruist parent,
therefore, maximizes his/her utility function, subject to budget constraint.
Others have suggest that the household consensus reflects that preferences
of a dominant family member (Ermisch, 1993).
Common Preference: An evaluation
Recent empirical evidence, however, does not support the 'common preference'
approach to household decision-making. The best evidence for the rejection
of income pooling hypothesis comes from a natural experiment in which some
husbands and wives received an exogenous income change as a result of the
change in UK child-allowance laws (Lundberg et al, 1997). In the late 1970s,
'universal child benefit,' which had consisted primarily of reduction in
the amount withheld for taxes from father's paycheck, was replaced by a
cash payment to the mother. The exogenous income change represented a substantial
redistribution (£ 500 or about 8% of average male earnings) of income
from fathers to mothers. Under the pooling hypothesis, this change should
have no effect on husbands and wives expenditure patterns. There was, however,
a "substantial increase" in spending on women's and children's
clothing, relative to men's clothing, following the policy change (Lundberg
et al, 1997). Holding total family income constant, it was found that the
income received by each spouse has substantial and significant effects on
family expenditure patterns (Lundberg et al, 1997).
The study of household risk pooling in an agrarian society also rejects
the notion of households as a single economic actor. Empirical studies of
detailed household survey data from Ghana reveals the complexities of intrahousehold
decision making. To investigate whether household members pool risk to smoothen
income shocks, the study examined whether the transitory incomes of individual
household members - as opposed to their permanent incomes - affect household
expenditure patterns. The results indicate that transitory income is spent
differently, depending on which household members accrue it, implying that
household members do no engage in full risk pooling - income smoothing -
with each other (Doss, 2001). Furthermore, men's and women's risk sharing
networks differ (Doss, 2001). If household do not pool income and risks
or act as a single economic agent, as the studies discussed above prove,
then how do they make their decisions?
Bargaining
Family members, like other economic actors, have varying tastes and preferences
and control different amounts of income. A more viable model of household
decision-making must therefore relax the common preferences and income pooling
assumption. Recent studies suggest that household members bargain to allocate
resources. Consider a typical household that consists of only two members
- a husband and wife. Each has a utility function that depends on his or
her consumption of private goods. According to the bargaining model, both
husband had wife have an input in decision-making (Gray, 1998). The weight
given to each persons input depends upon his or her opportunities outside
the family. The utility received by husband or wife from the outcome of
the bargaining, therefore, depends upon their utility outside marriage (Lundberg
et al, 1997). The higher one's utility outside marriage, the higher one's
utility in the outcome of the bargaining solution and lower the 'threat
point' threshold.
How the 'threat point' is resolved by husband and wife depends on the maximum
level of utility attainable outside marriage. If the utility received outside
marriage for either spouse is higher than the utility within a marriage,
there will be 'divorce threat' bargaining. In divorce-threat bargaining,
household allocation of resources will depend not on total family income
but on the income received by the husband and the income received by the
wife (Lundberg et al, 1997). The divorce threat point will also be affected
by exogenous factors, such as the conditions in the remarriage market and
the income available to divorced men and women. The intra-family allocation
of resources, according to the 'divorce threat bargaining model,' depends
on individual income and these exogenous factors.
The bargaining for household decision-making does not necessarily have to
be through an external 'divorce threat.' In most cases, the threat point
is internal to a marriage - the utility derived by the husband and the wife
in a non-cooperative marriage. Couples, in most cases, would prefer a non-cooperative
marriage to a divorce, since each spouse receives some benefits due to joint
consumption of public goods, such as housing. The husband and wife settle
their differences by Nash bargaining, but the alternative to a settlement
is an inefficient non-cooperative equilibrium within marriage. Each spouse
voluntarily provides household public goods, choosing utility maximizing
actions, given the actions of their partner (Lundberg et al, 1997). The
presence of internal threat point - non-cooperative marriage - has important
implications for household decision making. It generates family demand that
depend not on who receives income after divorce, like in 'divorce threat'
bargaining, but on who receives income within the marriage (Lundberg et
al, 1997). The family allocation of resources will therefore depend upon
the income independently received by the husband and by the wife. Each spouse,
in a non-cooperative marriage, treats the level of public good given by
the other spouse as fixed and chooses the quantity of his or her private
good and the public good to satisfy his or her utility (Lundberg et al,
1997).
Bargaining: An Evaluation
The family has been undergoing dramatic changes in recent year with changes
in social norms and laws regarding divorce and child support, allowing us
to examine the bargaining model for household decision-making. Typically,
literature on intra-household resource allocation has been constrained by
the difficulty of identifying within consumption surveys goods that are
clearly public goods for the household. One good that is usually a public
good is the family's charitable contribution. A survey designed to learn
about decision-making regarding household charitable giving helps understand
intra-household resource allocation. The study finds that men and women
have different preference for charitable giving. The study also concludes
that the differences in preferences are resolved through bargaining which
reduces charitable giving on an average by 6% (Andreoni et al, 2003).
A different way to study intrahousehold decision-making is to investigate
decisions which affect every member of the household differently, such as
decision to move to another part of the city or the country. Since it involves
changes in commuting times, neighborhood amenities and social network, a
decision to move has different effects on each member of the household differently.
A study examined the relationship between social and economic structure
and mobility decisions to determine if common preference or Nash bargaining
approach was more appropriate for modeling mobility. It concluded that mobility
behaviour differs with household's social and economic hierarchy, number
of generations in a household and the type of family (Chang et al, 2002).
The conclusion suggests that bargaining plays an important role in household
decision-making (Chang et al, 2002).
An interesting outcome of intrahousehold bargaining is the gender wage gap
(Elul et al, 2001). Gender wage gap results more from distribution across
occupations than from systematic discrimination against women (Elul et al,
2001). Empirical studies have shown that the gender gap is greatest between
married men and women. The difference in earnings is a consequence of men
tending to marry younger women. Marriage limits women's location and labour
mobility and hence their average earnings. In developed countries, women
are free not to marry and do so only if it is in their interests. A women's
decision of co-location with her husband, sometimes at the cost of lower
average income, is therefore the result of non cooperative bargaining (Elul
et al, 2001).
Conclusion
Households have traditionally been considered as a single economic actor
having a single utility function. Household decision making has been rationalized
as an attempt to maximize utility, subject to a budget constraint. The income
pooling assumption implies that ownership of income has no effect on household
allocation of resources. Empirical studies have, however, disproved the
'common preference' model of household decision-making and have shown that
men and women tend to have different preferences. Men and women tend to
respond differently to exogenous changes in incomes, suggesting some sort
of bargaining in household decision making. Most empirical studies have
proved the existence of both cooperative and non-cooperative bargaining
in household decision making. Whether the bargaining is cooperative or non-cooperative
depends on a spouses' utility within and outside the marriage. The bargaining
model of household decision making has important implications for public
policy and must be accounted for in redistribution and welfare public policy
decisions.
Andreoni, James; Brown, Eleanor; Rischall, Isaac (2003), "Charitable
Giving by Married Couples: Who Decides and Why Does It Matter?", Journal
of Human Resources v38, n1 (Winter 2003): 111-133.
Chang, Chin-Oh; Chen, Shu-Mei; Somerville, Tsur (2003), "Economic and
Social Status in Household Decision-Making: Evidence Relating to Extended
Family Mobility", in Urban Studies v40, n4 (April 2003): 733-746.
Doss, Cheryl R. (2001), "Is Risk Fully Pooled within the Household?
Evidence from Ghana", Economic Development and Cultural Change
v50, n1 (October 2001): 101-130.
Ermisch, John, (1993), "Familia Oeconomica: A Survey of the Economics
of the Family", in Scottish Journal of Political Economy, Vol. 40,
No. 4, November 1993, pp. 353-374.
Elul, Ronel; Silva-Reus, Jose; Volij, Oscar (2002), "Will You Marry
Me? A Perspective on the Gender Gap", Journal of Economic Behavior
and Organization v49, n4 (December 2002): 549-572.
Gray, Jeffrey S., (1998), "Divorce-Law Changes, Household Bargaining,
And Married Women's Labor Supply", The American Economic Review,
Vol. 88, No. 3, June 1998, Vol. 88, No. 3, pp. 628-642.
Klawon, Emily; Tiefenthaler, Jill , (2001), "Bargaining over Family
Size: The Determinants of Fertility in Brazil", Population Research
and Policy Review v20, n5 (October 2001): 423-440.
Lundberg, Shelly and Robert A. Pollak, (1996), "Bargaining and Distribution
in Marriage", in The Journal of Economic Perspectives, Vol.
10, No. 4, Fall 1996, pp. 139-158.
Lundberg, Shelley; Robert Pollak and Terrance Wales, (1997), "Do Husbands
and Wives Pool Their Resources?", The Journal of Human Resources,
Vol. 32, No. 3, 1997, pp. 463-479.
Lundberg, Shelly; Startz, Richard; Stillman, Steven (2003), "The Retirement-Consumption
Puzzle: A Marital Bargaining Approach", in Journal of Public Economics
v87, n5-6 (May 2003): 1199-1218.